Ce mémoire s'inscrit dans la construction d'une méthodologie d'analyse de défaillance pour l'évaluation de la fiabilité de diodes électroluminescentes, par une approche basée sur l'analyse physique de dégradation et l'extraction de signatures électriques de défaillance et optiques pour localiser les zones dégradées. L'ajout d'analyses physico-chimiques réduit le nombre de composants et peut confirmer les mécanismes de dégradation induits par les vieillissements en stockage actif. Un projet, en collaboration avec le CNES, a permis la mise en évidence des zones sensibles de DELs à MPQ InGaN/GaN à faible puissance (30mW) soumises à un vieillissement en conditions opérationnelles (1500h/85°C/Inominal). L'analyse de défaillance de ces DELs a permis d'expliquer une perte de 65% de puissance optique par la modification de la structure moléculaire de l'huile silicone activée photothermiquement induisant une perte de fluorescence de 69% et une très forte diminution de l'absorption de la lumière de la DEL (90%). Nous avons également démontré (projet CEA-LETI - éclairage public) que le même mécanisme est présent dans le mélange gel silicone/phosphore YAG:Ce de DELs blanches à MPQ InGaN/GaN soumises à un vieillissement similaire (85°C/550mA/500h). A 450nm, le rendement de fluorescence a augmenté de 1,2% malgré des pertes en absorption (> 94%) et en réémission de fluorescence (> 85%). La modification de la structure moléculaire du gel a induit une perte de puissance optique des DELs de 45% et une dérive de la couleur blanche vers le jaune (≈ 3,6%). Cette dérive est due à un décalage spectral de la fluorescence de l'UV (5nm) vers le bleu entraînant un décalage vers le rouge (2nm) de la lumière de la DEL.

Ce travail concerne l'étude du phénomène de décharges partielles dans les matériaux isolants utilisés en électronique de puissance. En utilisant des méthodes de détection électrique et optique, le mécanisme de décharge partielle sur des substrats d'AlN dans l'huile silicone a été étudié sur un grand nombre d'échantillons. La variation de la nature du substrat (AlN, Al2O3 et composite verre/époxy) et du matériau d'encapsulation (huile silicone, huile de colza, huile minérale de transformateur, liquide d'imprégnant du condensateur Jarylec et Ugilec) met en évidence l'origine des décharges partielles de l'ensemble substrat - encapsulant. Les décharges partielles sur les substrats céramiques frittés ne dépendent pas du passivant, et se produisent dans le volume du substrat. L'évolution temporelle de la lumière émise dans les liquides en configuration pointe - plan et sur le substrat dans différents liquides montre que l'émission de lumière est un phénomène très complexe influencé par de nombreux paramètres : électroluminescence du solide, de l'encapsulant, décharges partielles, absorption des matériaux. Le phénomène d'électroluminescence du liquide est activé par une illumination extérieure. Les mesures de spectroscopie diélectrique haute tension n'apportent pas d'information supplémentaire sur le phénomène de décharges partielles, car les pertes correspondantes sont très faibles. Mots clés : Décharge partielle, électroluminescence, Nitrure d'aluminium, Alumine, huile silicone, diélectrique liquide, électronique de puissance.

Background A perforation in the nasal septum can cause symptoms such as bleeding, obstruction, crusts and pain, and can be a challenge to treat. Surgery is the treatment of choice, but disease, size of the perforation, or the patient’s wish may contradict surgery. A custom-made silicone obturator is a successful treatment option, but little is known how this treatment affects the microbial flora. The purposes of this thesis were (i) to investigate the microbial flora around symptomatic nasal septal perforations before treatment, (ii) during and after a 12-month treatment period with a custom-made obturator, (iii) to compare the microbial flora around symptomatic perforations with the flora from the same area of the septum in healthy individuals, (iv) to investigate the microbial colonization of the silicone obturator, and (v) also to investigate the water sorption, solubility and if the wettability of silicones are affected by water. The hypotheses were (i) that the bacterial flora around symptomatic perforations would not differ from that found in healthy individuals, apart from a possible presence of Helicobacter pylori; (ii) the bacterial flora would change in composition during the course of treatment and that microorganisms and proteins could be seen on the surface of the silicone obturators; (iii) a material that has adsorbed water would also show an increase in wettability and the surface free energy of the material.  Methods Twenty-seven patients and 101 healthy individuals volunteered. Swabs were made around the rim of the perforation, or on the septum in the locus Kisselbachi area in the healthy individuals. Bacteria and fungi were isolated and identified with standard laboratory techniques. A biopsy of the granulated tissue at the perforation was taken and cultivated for Helicobacter pylori. Swabs were also taken three, six and twelve months after inserting the obturator. The obturator was analysed after being used twelve months in the nose.  Seven silicones were tested for water sorption and solubility according to ISO standards 1567:1999 and ISO 10477:2004. The change in wettability was examined by measuring the contact angle with a contact goniometer at various stages of the sorption/solubility test. Results Staphylococcus aureus was present in 88% of the untreated patients. With treatment a significant reduction of S. aureus occurred to 54.5% (p<0.05). In the healthy group S. aureus was present in 13% of the subjects. No Helicobacter pylori could be cultivated from the biopsies taken of the granulated tissue at the perforation. The flora round the untreated perforation was dominated by S. aureus with few other bacterial species detected. In the healthy group there was a diversified flora with both aerobic and anaerobic bacteria. SEM revealed a rough surface on the silicone obturator and crazing of the silicone surrounding the pigment granules. Both bacteria and proteins could be seen on the obturators in SEM. Candida albicans was detected in one obturator, but not in the mucosal swab at the corresponding time. That patient had, however, been treated for Candida in the nose six months prior to the last visit in the study. Wettability was affected but did not increase with amount of adsorbed water. Some materials showed an increase and some a decrease in the surface-free energy. The tested addition silicones showed little sorption and solubility. Conclusions The patients with symptomatic perforations of the nasal septum had a bacterial flora totally dominated by S. aureus. The massive presence of S. aureus around symptomatic perforations may have an impact on the persistence of the granulated and inflamed tissue present in symptomatic perforations, thus forming a vicious circle with bleeding and crustation. S. aureus dominance in the mucosa surrounding symptomatic perforations was diminished by using a custom-made obturator. The microbial flora became more diversified with the treatment, although not resembling the flora in healthy individuals. The microbial flora of the obturators was similar, but not the same as the corresponding mucosal flora. The discovery of Candida in the obturator of a patient who had been treated for Candida in the nose six months earlier suggests that obturators need to be exchanged when fungal infections are being treated to prevent the fungus from re-infecting the patient at a later stage. The silicone had a rough surface and a poor wettability, both aspects favours colonization of microorganisms. The silicone was negatively affected by the colouring pigments, this should be considered when colouring is not necessary. The slight, but existing solubility of silicones emphasises the importance of using medical grade silicones that are more purified than industrial silicones.

Silicone rubber based on polydimethylsiloxane is used ashigh voltage outdoor insulation, due to its ability to preservethe hydrophobic surface properties during service and evenregain hydrophobicity after exposure to electrical discharges.The underlying processes for the hydrophobic recovery arediffusion of low molar mass siloxanes from the bulk to thesurface and reorientation by conformational changes ofmolecules in the surface region. Only little is known of whichfactors are responsible for the long-term stability of thishydrophobic recovery. It is therefore important to increase theknowledge about the fundamental mechanisms for the loss andrecovery of hydrophobicity of silicone rubbers, exposed toelectrical discharges. Addition-cured polydimethylsiloxanenetworks, with known crosslink densities, were exposed tocorona discharges and air/oxygen-plasma and the loss andrecovery of hydrophobicity was characterised by contact anglemeasurements. The degree of surface oxidation increased withincreasing exposure time with a limiting depth of 100- 150 nm,as assessed by neutron reflectivity measurements. The oxidationrate increased with increasing crosslink density of the polymernetwork, according to X-ray photoelectron spectroscopy. Withinthe oxidised layer, a brittle, silica-like layer was graduallydeveloped with increasing exposure time. The hydrophobicrecovery following the corona or air/oxygen- plasma exposuresoccurred at a slow pace by diffusion of cyclic oligomericdimethylsiloxanes through the micro-porous but uncrackedsilica-like surface layer or at a much higher pace by transportof the oligomers through cracks in the silica-like layer. Theoligomers were present in the bulk, but additional amounts wereformed during exposure to corona discharges. High-temperaturevulcanised silicone rubber specimens were aged in a coastalenvironment under high electrical stress levels (100 V/mm). Thechanges in surface structure and properties were compared tothe data obtained from specimens exposed to coronadischarges/plasma. The dominating degradation mechanism wasthermal depolymerisation, initiated by hot discharges. Thisresulted in the formation of mobile siloxanes, of which the lowmolar mass fraction consisted of cyclic oligomericdimethylsiloxanes. Oxidative crosslinking resulting insilica-like surface layers was not observed during theseconditions. <b>Keywords:</b>silicone rubber, polydimethylsiloxane,hydrophobicity, corona, air-plasma, oxygen-plasma, surfacecharacterisation, degradation products, crosslink density.

A clutch actuator is used in a vehicle to transmit movement and force from the clutch pedal to the release bearing of the clutch. A pneumatic clutch actuator consists of an anodised aluminium cylinder, inside of which a piston, with a rubber lip seal and a PTFE guiding ring, slides. The system is lubricated with silicone grease before assembly. A commercial clutch actuator of this type, has a service life of 3 million actuations and must function in a wide temperature range, from -40 ºC to 140 ºC. In this thesis, the complex tribological system of pneumatic clutch actuators has been studied. Field worn actuators have been disassembled and investigated. A laboratory test method has been developed to understand the tribomechanisms present in pneumatic clutch actuators. The test method's capability of simulating the real contact has been verified, by the comparison with studied actuators from the field. The influence of contact parameters: temperature, load, lubrication and particle contamination, has been investigated. In addition, different anodised aluminium surfaces have been studied. The manufacturing method of the aluminium cylinder influences surface topography and structure of the oxide, resulting in different mechanical and frictional properties. The wear during tests with only silicone grease is reminiscent, but on a lower scale, to the wear during tests with a mixture of silicone grease and standard dust. The initially applied amount of silicone grease is not important, the friction seems to depend on the amount of silicone grease that is dragged into or pushed out from the contact area during testing. Silicone grease lubrication reduces wear of the lip seal. However, during some tests, an adhesive layer, composed of  grease residuals and some PTFE, was formed on the lip. A triple PTFE transfer, from guiding ring to aluminium surface, to lip seal, to aluminium surface, occurred. Such transfer of material from the PTFE guiding ring was detected from the unlubricated tests, and also from the silicone grease lubricated tests, i.e. silicone grease lubrication does not prevent PTFE material transfer.

The wettability of contact lenses has become an area of intense research, with the belief that the more "hydrophilic" or wettable the lens surface is, the more comfortable the lens may be, as the posterior surface of the eyelid will move more smoothly over it, hence increasing comfort. <br /><br /> There are many ways to assess the wettability of a given material, namely sessile drop,¹ captive bubble ² or Wilhelmy plate. ³ This thesis used the sessile drop method to determine the surface wettability of various hydrogel contact lens materials, by measuring the advancing contact angle made between the lens surface and a pre-determined volume of HPLC-grade water. This was followed by measuring the surface wettability following periods in which the lens materials were soaked in various contact lens care regimens. Further studies determined wettability of lens materials after various periods of in-eye wear and finally a study was undertaken to evaluate if a novel biological technique could be used to differentiate proteins that deposit on hydrogel lens materials that may affect wettability and cause discomfort. <br /><br /> A variety of hydrogel lenses, taken directly from their packaging and after soaking in various care regimens, were analyzed to determine their sessile drop advancing contact angles, in vitro. These studies indicated that poly-2-hydroxyethylmethacrylate (pHEMA)-based lenses are inherently more wettable than silicone-based lenses, unless they have a surface treatment that completely covers the hydrophobic siloxane groups. Additionally, certain combinations of lens materials and care regimens produce inherently more wettable surfaces when measured in vitro. <br /><br /> Suitable methods to assess contact lens wettability ex vivo, or after subjects had worn lenses for set periods of time, were developed. It was determined that using latex gloves to remove lenses had no impact upon the lens surface wettability and that rinsing of the lens surface after removal from the eye was required to determine the wettability of the underlying polymer. <br /><br /> The final wettability studies involved an analysis of various lens materials from clinical studies conducted within the Centre for Contact Lens Research (CCLR). These studies investigated differences in wettability between silicone hydrogel lenses manufactured from differing polymers and variations in ex vivo wettability of several combinations of lens materials and solutions, worn for varying periods of time. <br /><br /> A novel method to investigate proteins extracted from lenses using 2D-Difference in Gel Electrophoresis (DIGE) found that this technique could be used to analyze proteins extracted from contact lenses. The data obtained showed that there was no difference between a group of subjects who were symptomatic of lens-induced dryness or a control group, and that care solutions had a minimal influence on the pattern of deposition seen. <br /><br /> The overall conclusion of these studies is that hydrogel lens wettability is affected by the polymer composition and that care regimen components can modify the surface wettability.

The wettability of contact lenses has become an area of intense research, with the belief that the more "hydrophilic" or wettable the lens surface is, the more comfortable the lens may be, as the posterior surface of the eyelid will move more smoothly over it, hence increasing comfort. <br /><br /> There are many ways to assess the wettability of a given material, namely sessile drop,¹ captive bubble ² or Wilhelmy plate. ³ This thesis used the sessile drop method to determine the surface wettability of various hydrogel contact lens materials, by measuring the advancing contact angle made between the lens surface and a pre-determined volume of HPLC-grade water. This was followed by measuring the surface wettability following periods in which the lens materials were soaked in various contact lens care regimens. Further studies determined wettability of lens materials after various periods of in-eye wear and finally a study was undertaken to evaluate if a novel biological technique could be used to differentiate proteins that deposit on hydrogel lens materials that may affect wettability and cause discomfort. <br /><br /> A variety of hydrogel lenses, taken directly from their packaging and after soaking in various care regimens, were analyzed to determine their sessile drop advancing contact angles, in vitro. These studies indicated that poly-2-hydroxyethylmethacrylate (pHEMA)-based lenses are inherently more wettable than silicone-based lenses, unless they have a surface treatment that completely covers the hydrophobic siloxane groups. Additionally, certain combinations of lens materials and care regimens produce inherently more wettable surfaces when measured in vitro. <br /><br /> Suitable methods to assess contact lens wettability ex vivo, or after subjects had worn lenses for set periods of time, were developed. It was determined that using latex gloves to remove lenses had no impact upon the lens surface wettability and that rinsing of the lens surface after removal from the eye was required to determine the wettability of the underlying polymer. <br /><br /> The final wettability studies involved an analysis of various lens materials from clinical studies conducted within the Centre for Contact Lens Research (CCLR). These studies investigated differences in wettability between silicone hydrogel lenses manufactured from differing polymers and variations in ex vivo wettability of several combinations of lens materials and solutions, worn for varying periods of time. <br /><br /> A novel method to investigate proteins extracted from lenses using 2D-Difference in Gel Electrophoresis (DIGE) found that this technique could be used to analyze proteins extracted from contact lenses. The data obtained showed that there was no difference between a group of subjects who were symptomatic of lens-induced dryness or a control group, and that care solutions had a minimal influence on the pattern of deposition seen. <br /><br /> The overall conclusion of these studies is that hydrogel lens wettability is affected by the polymer composition and that care regimen components can modify the surface wettability.

The temperature and thermal stress distributions and variations of the high power LED module were studied in this work. The thermal-elastic-plastic 3D finite element models of MSC.marc software package are employed to simulate these performances for the high power LED module. Two high power white light LED module designs are investigated¡G one is the traditional phosphorescent silicone with blue LED module and the other is a phosphor glass lens with blue LED module. The distributions of temperature and thermal stress of in these two operating LED modules are compared and discussed. The effects of different packaging parameters¡Ge.g. bonding materials, substrate materials, lens materials on the temperature and thermal stress have also been studied in this work. The simulated results reveal that the serious thermal crack may occur for these two designs if the power of single die is over 10 watt. The simulated results also indicate that an attached fin cooler may improve these thermal crack disadvantaged significantly. The effect of fin design parameters on the peak temperature reduction has studied. A feasible fin design for the high power LED module has also been proposed.

<p>Silicone rubber based on polydimethylsiloxane is used ashigh voltage outdoor insulation, due to its ability to preservethe hydrophobic surface properties during service and evenregain hydrophobicity after exposure to electrical discharges.The underlying processes for the hydrophobic recovery arediffusion of low molar mass siloxanes from the bulk to thesurface and reorientation by conformational changes ofmolecules in the surface region. Only little is known of whichfactors are responsible for the long-term stability of thishydrophobic recovery. It is therefore important to increase theknowledge about the fundamental mechanisms for the loss andrecovery of hydrophobicity of silicone rubbers, exposed toelectrical discharges. Addition-cured polydimethylsiloxanenetworks, with known crosslink densities, were exposed tocorona discharges and air/oxygen-plasma and the loss andrecovery of hydrophobicity was characterised by contact anglemeasurements. The degree of surface oxidation increased withincreasing exposure time with a limiting depth of 100- 150 nm,as assessed by neutron reflectivity measurements. The oxidationrate increased with increasing crosslink density of the polymernetwork, according to X-ray photoelectron spectroscopy. Withinthe oxidised layer, a brittle, silica-like layer was graduallydeveloped with increasing exposure time. The hydrophobicrecovery following the corona or air/oxygen- plasma exposuresoccurred at a slow pace by diffusion of cyclic oligomericdimethylsiloxanes through the micro-porous but uncrackedsilica-like surface layer or at a much higher pace by transportof the oligomers through cracks in the silica-like layer. Theoligomers were present in the bulk, but additional amounts wereformed during exposure to corona discharges. High-temperaturevulcanised silicone rubber specimens were aged in a coastalenvironment under high electrical stress levels (100 V/mm). Thechanges in surface structure and properties were compared tothe data obtained from specimens exposed to coronadischarges/plasma. The dominating degradation mechanism wasthermal depolymerisation, initiated by hot discharges. Thisresulted in the formation of mobile siloxanes, of which the lowmolar mass fraction consisted of cyclic oligomericdimethylsiloxanes. Oxidative crosslinking resulting insilica-like surface layers was not observed during theseconditions.</p><p><b>Keywords:</b>silicone rubber, polydimethylsiloxane,hydrophobicity, corona, air-plasma, oxygen-plasma, surfacecharacterisation, degradation products, crosslink density.</p>

Microbial colonization on the surface of silicone rubber high voltage outdoor insulators often results in the formation of highly hydrated biofilm that influence the surface properties, such as surface hydrophobicity. The loss of hydrophobicity might lead to dry band formation, and, in the worst cases, flashover and failure of the insulator. In this work, the biocidal effects of various antimicrobial compounds in silicone rubber materials were determined. These materials were evaluated according to an ISO standard for the antimicrobial activity against the growth of aggressive fungal strains, and microorganisms that have been found colonizing the surfaces of outdoor insulators in several areas in the world. Several compounds suppressed microbial growth on the surfaces of the materials without compromising the material properties of the silicone rubber. A commercial biocide and thymol were very effective against fungal growth, and sodium benzoate could suppress the fungal growth to some extent. Thymol could also inhibit algal growth. However, methods for preservation of the antimicrobial agents in the bulk of the material need to be further developed to prevent the loss of the compounds during manufacturing. Biofilm formation affected the surface hydrophobicity and complete removal of the biofilm was not achieved through cleaning. Surface analysis confirmed that traces of microorganisms were still present after cleaning. Further, surface modification of the silicone rubber was carried out to study how the texture and roughness of the surface affect biofilm formation. Silicone rubber surfaces with regular geometrical patterns were evaluated to determine the influence of the surface texture on the extent of microbial growth in comparison with plane silicone rubber surfaces. Silicone rubber nanocomposite surfaces, prepared using a spray-deposition method that applied hydrophilic and hydrophobic nanoparticles to obtain hierarchical structures, were studied to determine the effects of the surface roughness and improved hydrophobicity on the microbial attachment. Microenvironment chambers were used for the determination of microbial growth on different modified surfaces under conditions that mimic those of the insulators in their outdoor environments. Different parts of the insulators were represented by placing the samples vertically and inclined. The microbial growth on the surfaces of the textured samples was evenly distributed throughout the surfaces because of the uniform distribution of the water between the gaps of the regular structures on the surfaces. Microbial growth was not observed on the inclined and vertical nanocomposite surfaces due to the higher surface roughness and improved surface hydrophobicity, whereas non-coated samples were colonized by microorganisms. / <p>QC 20151002</p>